changes

fusion_accounting/fusion_accounting_ocr/services/attachment_to_image.py

@@ -0,0 +1,43 @@
+"""Helper: turn an ir.attachment into a list of PIL.Image pages.
+
+Kept separate from the adapters so future backends (Ollama-vision, Mindee)
+that want PIL images directly don't have to re-implement the PDF rendering.
+"""
+
+import base64
+import io
+import logging
+
+_logger = logging.getLogger(__name__)
+
+
+def attachment_to_pages(attachment):
+    """Decode an ir.attachment into a list of PIL.Image pages.
+
+    Returns ``[]`` on failure (caller should treat as no pages).
+    """
+    try:
+        from PIL import Image
+        from pdf2image import convert_from_bytes
+    except ImportError as e:
+        _logger.warning("attachment_to_pages requires PIL + pdf2image: %s", e)
+        return []
+
+    if not attachment or not attachment.datas:
+        return []
+
+    try:
+        data = base64.b64decode(attachment.datas)
+    except Exception as e:
+        _logger.warning("Could not decode attachment %s: %s", attachment.id, e)
+        return []
+
+    mimetype = attachment.mimetype or ''
+    is_pdf = mimetype == 'application/pdf' or data[:4] == b'%PDF'
+    try:
+        if is_pdf:
+            return convert_from_bytes(data, dpi=200)
+        return [Image.open(io.BytesIO(data))]
+    except Exception as e:
+        _logger.warning("Could not render attachment %s: %s", attachment.id, e)
+        return []